|
From: | Anthony Liguori |
Subject: | Re: [Qemu-devel] Re: [PATCH 15/15] Implement the bus structure for PAPR virtual IO |
Date: | Sun, 13 Feb 2011 09:14:10 -0600 |
User-agent: | Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.15) Gecko/20101027 Lightning/1.0b1 Thunderbird/3.0.10 |
On 02/13/2011 06:40 AM, Alexander Graf wrote:
Ah, yeah. I'm still not sure what to do about it. I was going to fold the dynamic hcall registration into the patch set before upstreaming. But then something paulus said made me rethink whether the dynamic registration was a good idea. Still need to sort this out before the series is really ready.We can surely move it to dynamic later on. I think the "proper" way would be to populate a qdev bus and have the individual hypercall receivers register themselves through -device creations.
Ignore the qdev bit for a moment. Hypercalls could be plausibly implemented as another I/O space from a processor so the thing to model off of would be PIO dispatch (cpu_outb and friends).
From a qdev perspective, having a VIO bus makes sense. The details of which I/O spaces are uses are not as important from a device tree perspective.
Regards, Anthony Liguori
Alex
[Prev in Thread] | Current Thread | [Next in Thread] |